A few weeks ago, Ravencoin, underwent a successful proof of work algorithm upgrade. Congratulations to their community and all the hard work put in to make the transition as smooth as possible. It looks like everything went smoothly and the network is moving along nicely.
One of the design goals of the Ravencoin project was to prevent large mining pools and ASICs from dominating the network and instead encourage a widely dispersed mining ecosystem with commodity hardware CPU and GPUs to emerge.
In the very early stages of Bitcoin’s development, Bitcoin could be mined on commodity hardware but as the network grew, mining became more professionalized, industrialized, and centralized because of the development of specialized hardware (ASICs). ASICs used for Bitcoin mining are about 1000x more efficient when compared to mining with common CPUs and GPUs. In the mining context, efficiency means better performance per watt and/or superior performance per dollar.
The Ravencoin community wanted to at least give its network time to mature and cultivate an active, dispersed, hobbyist community of miners that could use their CPUs and/or GPUs to contribute to the mining process. They initially used the x16r mining algorithm and have been somewhat successful in cultivating a network of commodity hardware miners but the development and mining communities felt the need to improve the network’s ability to resist domination by ASICs.
Keep in mind that ASIC resistance does not mean, complete deterrence of ASICs or the complete incompatibility of ASICs. ASICs can still be used for mining on “ASIC resistant” mining algorithms. However their advantages are severely diminished with the use of certain techniques. The performance gaps between ASICs and commodity hardware are considerably compressed when using effective ASIC resistant mining algorithms.
The network officially move away from the x16r mining algorithm to a new mining algorithm named, KAWPOW, The expectation from this change is that most modern GPUs will be on a level playing field with ASICs when mining Ravencoin. This allows anyone with a decent computer to run Ravencoin mining software successfully.
The KAWPOW algorithm is a tweaked version of the Programmatic Proof of Work (ProgPoW) algorithm developed by Ethereum developers trying to improve Ethash’s (Ethereum’s current PoW algo) ASIC resistance. ProgPoW, and KAWPOW in the case of Ravencoin, is designed to programmatically reconfigure itself at a regular interval in such a way as to make the development of an ASIC impractical or economically unmotivating. A moving, dynamic target is harder to hit than a static target of similar size.
Developers of ProgPoW and by extension developers of KAWPOW designed these algorithms to make full use of the programmability and capabilities of all the hardware on a modern GPU. Maximizing the use of the entirety of a GPU minimizes the gains to be won by developing an ASIC considerably, adding to the network’s ASIC “resistance”. This doesn’t mean ASICs are impossible to produce for ProgPow and KAWPOW, but they are much less cost-effective for the performance they offer over a GPU.
Other Points of View
Many remain skeptical of the efficacy of ASIC resistance. Although it does lower the barrier to entry compared to ASIC dominated networks, this doesn’t necessarily imply increased security. A few networks that have optimized for ASIC resistance, such as VTC, BTG, and XVG, have been successfully attacked. We have not seen this happen to networks with specialized hardware dominating mining.
Networks that favor mining with widely available hardware like CPUs and GPUs open themselves to attack from a wide range of computing power that has no connection or commitment to the network. This computing power can attack the network and then move on to other operations with little economic downside.
Networks that have specific hardware dedicated to mining, like Bitcoin, don’t have this problem. There are no computers outside of the network itself that can attack it. The attacker wouldn’t have the appropriate hardware. An outside party would have to develop or buy a lot of specific hardware to attack a network secured by ASICs. It is true, miners in this type of network can attack the network they’re mining, but they’re destroying the only use of their hardware as well as faith in a network they’ve made a considerable investment in.This would be extremely expensive and highly irrational considering there are few, if any, other uses for their equipment.
However, hardware for ASIC secured networks are typically produced by only a handful of manufacturers due to economies of scale. This concentration of manufacturing can make many nervous about a network’s vulnerabilities and dependence on a few entities. A network dominated by general purpose hardware doesn’t have to contend with this issue.
It’s early in the history of these networks, but it looks like a multiplicity of security architectures could conceivably co-exist and provide good enough security for their respective networks. Regardless of the setup, the security architecture of all of these networks depends on the assumption that honest nodes will outnumber and outmaneuver malicious nodes. If this stays true for Ravencoin, and they are able to adjust to threats successfully like they’ve shown the ability to do, the network has a bright future.
It’ll be interesting to see how effective KAWPOW can be at fending off ASICs and what the Ravencoin community will do to combat this ever looming threat.